Representations for Genetic and Evolutionary Algorithms

Franz Rothlauf

ISBN 10: 3642064108 / ISBN 13: 9783642064104

Editore: Springer Berlin Heidelberg, 2010

Riguardo questo articolo

Druck auf Anfrage Neuware - Printed after ordering - In the field of genetic and evolutionary algorithms (GEAs), a large amount of theory and empirical study has been focused on operators and test problems, while problem representation has often been taken as given. This book breaks with this tradition and provides a comprehensive overview on the influence of problem representations on GEA performance. The book summarizes existing knowledge regarding problem representations and describes how basic properties of representations, such as redundancy, scaling, or locality, influence the performance of GEAs and other heuristic optimization methods. Using the developed theory, representations can be analyzed and designed in a theory-guided matter. The theoretical concepts are used for solving integer optimization problems and network design problems more efficiently. The book is written in an easy-readable style and is intended for researchers, practitioners, and students who want to learn about representations. This second edition extends the analysis of the basic properties of representations and introduces a new chapter on the analysis of direct representations. Codice articolo 9783642064104

Dati bibliografici

Titolo: Representations for Genetic and Evolutionary...

Casa editrice: Springer Berlin Heidelberg

Data di pubblicazione: 2010

Legatura: Taschenbuch

Condizione: Neu

Descrizione articolo

Riassunto:

In the field of genetic and evolutionary algorithms (GEAs), a large amount of theory and empirical study has been focused on operators and test problems, while problem representation has often been taken as given. This book breaks with this tradition and provides a comprehensive overview on the influence of problem representations on GEA performance. The book summarizes existing knowledge regarding problem representations and describes how basic properties of representations, such as redundancy, scaling, or locality, influence the performance of GEAs and other heuristic optimization methods. Using the developed theory, representations can be analyzed and designed in a theory-guided matter. The theoretical concepts are used for solving integer optimization problems and network design problems more efficiently. The book is written in an easy-readable style and is intended for researchers, practitioners, and students who want to learn about representations. This second edition extends the analysis of the basic properties of representations and introduces a new chapter on the analysis of direct representations.

Contenuti:

Representations for Genetic and Evolutionary Algorithms.- Three Elements of a Theory of Representations.- Time-Quality Framework for a Theory-Based Analysis and Design of Representations.- Analysis of Binary Representations of Integers.- Analysis and Design of Representations for Trees.- Analysis and Design of Search Operators for Trees.- Performance of Genetic and Evolutionary Algorithms on Tree Problems.- Summary and Conclusions.